Risk is an inherent part of health care, particularly in large referral centers, where some of the most complex cases are managed. While risk cannot be eliminated from the clinical activities, it is believed that some practices involving unnecessary risk can be mitigated without impacting overall performance. Up until now, our ability to identify these vulnerable practices, and develop durable preventive or mitigating strategies, was hampered by outdated models of risk and an inadequate approach to the analysis of risk. In an effort to develop more realistic models of risk in complex health care settings, we applied a system dynamics framework to model how features of the environment (e.g., time pressures, resource shortages, etc.) and human attributes (e.g., risk tolerance, confidence in existing safety policies, etc.) combine to influence safety. In particular, we studied the complex interactions between production pressures, historical experience with adverse outcomes, inherent risk tolerance/propensity, confidence in and compliance with safety controls in the ambulatory procedural domain. We also studied how these interactive factors drive the system above acceptable thresholds of safety. Using prospective modeling and simulation, we found that a large number of safety controls relating to procedural sedation, staffing, and case scheduling had been "disengaged" or waived by physicians and nurses in the interest of maximizing the productivity and efficiency outcome objectives. These types of violations in most instances reflected a sound decision using a local judgment criterion given the time, resource constraints, throughput pressures, and short-term incentives that shape behavior. However, we were also able to demonstrate through simulation that if time pressures or delays could be relaxed or reduced through process redesign, many of these dysfunctional behaviors would dissipate. Based on this work, we have designed a series of interventions aimed at reducing the system pressures that negatively influence provider behavior and create unsafe operating conditions within the ambulatory procedural domain. In the work proposed here, we will implement these interventions and evaluate their impact on safety. [unreadable] [unreadable] PUBLIC HEALTH RELEVANCE: Risk is an inherent part of health care, particularly in large referral centers, where some of the most complex cases are managed. While risk cannot be eliminated from the clinical activities, it is believed that some practices involving unnecessary risk can be mitigated without impacting overall performance. We recently completed a prospective risk assessment of ambulatory procedural care in an attempt to identify critical features of this complex system that contributed to unnecessary risk. Based on our preliminary work, we believe that in order to optimize safety in this area of ambulatory procedural care, we must develop and implement: [unreadable] * A highly reliable screening tool to assess the need for anesthesia involvement in procedural sedation [unreadable] * A transformation of the booking/scheduling process that: [unreadable] o incorporates the screening tool, and minimizes opportunities for bypassing the screening process [unreadable] o enables realistic and balanced daily case loads for the unit measured in terms of case complexity and risk (concept of "balanced booking") [unreadable] o has sufficient flexibility or adaptiveness to meet emerging demand (i.e., urgent add-on cases) [unreadable] o enables reliable estimation of demand for anesthesia support [unreadable] o enables reliable estimation of overall case load for the day [unreadable] In this proposed work we will design, implement and evaluate these interventions, and develop a toolkit for use by other organizations to optimize safety in the ambulatory procedural domain. [unreadable] [unreadable] [unreadable]